多电机多包传输分布式控制系统故障检测优化策略  

Fault Detection Optimization Strategy for Multi-motor Distributed Control System with Multi-packet Transmission

作　　者：纪烨晴 郑焕坤[1] JI Yeqing;ZHENG Huankun(School of Electrical and Electronic Engineering,North China Electric Power University,Baoding 071003,China)

机构地区：[1]华北电力大学电气与电子工程学院

出　　处：《华北电力大学学报（自然科学版）》2019年第6期49-57,共9页Journal of North China Electric Power University：Natural Science Edition

基　　金：国家自然科学基金资助项目(51476187);国家电网公司科技资助项目(5228001700CW);中央高校基本科研业务费专项资金资助项目(2018QN074)

摘　　要：为同时提升故障检测的故障灵敏度和噪声鲁棒性,针对具有双通道多包传输的短时变时延多电机分布式控制系统,提出了一种基于改进自适应混沌果蝇优化算法的传感器故障检测观测器优化设计方法。首先将双通道多包传输转化为切换系统,短时变时延视为一种系统不确定性。然后将观测器残差信号对噪声信号和故障信号的传递函数比值作为适应度函数,通过改进自适应混沌果蝇优化算法对适应度函数进行优化得到最优观测器增益矩阵。最后将方法在分布式控制系统半物理平台上进行仿真验证,结果显示相比于传统的鲁棒故障检测方法来说,提出的优化方法能够同时抑制噪声信号和放大故障信号,从而提高了故障诊断的正确率,降低了虚警率。In order to enhance fault sensitivity and noise robustness in fault detection simultaneously,this paper proposed an optimal design method for sensor fault detection observer.The design method,which was based on improved adaptive chaotic fruit fly optimization algorithm,was targeted at distributed control system of multi-motors with short time-varying delays and multi-packets transmission.First step of the method was to convert dual channel multiple packet transmission into switched system.By the way,short time-varying delay was deemed as a systematic uncertainty.Next step was to obtain the optimal observer gain matrix by optimizing the fitness function via the algorithm mentioned above.The fitness function was the ratio of transfer function of the residual signal to the noise signal and fault signal as fitness function.Finally,this paper verified the method in semi-physical platform of distributed control system.The simulation results show that compared with traditional robust fault detection,the proposed method suppresses noise signal and amplifies fault signal,thereof effectively improving the accuracy of fault diagnosis and reducing false alarm rate.

关 键 词：多电机 分布式控制系统 自适应混沌果蝇算法 故障检测 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]